Crane04/iris

Face match system for medical purposes - Rust africa hackathon # Iris — Biometric Infrastructure **Iris** is a stateless, high-performance face recognition infrastructure built in Rust. It provides a REST API designed for hospital IT systems to identify unresponsive patients in real-time by comparing emergency captures against secure patient databases. --- ## Core Philosophy - **Stateless by Design:** Images are processed in RAM and destroyed immediately after feature extraction. No biometric data ever touches the disk. - **Infrastructure, Not Storage:** Iris does not store medical records. It returns a mathematical similarity score between two images, allowing hospitals to link to their own secure EMR systems. - **High Performance:** Powered by Rust and ONNX-accelerated models (YuNet and SFace) for sub-100ms inference. --- ## Prerequisites Before running Iris, ensure you have the following installed: - **Rust:** Install via rustup - **OpenCV 4.x:** \* _macOS:_ `brew install opencv` - _Linux:_ `sudo apt install libopencv-dev` - _Windows:_ Follow OpenCV-Rust installation guide --- ## Installation & Setup ### Clone the Repository ```bash git clone github.com cd iris ``` ## Run Web ```bash cd web npm install ``` ## Run API ```bash cd ../api ``` ## Download AI Models Iris requires pre-trained ONNX models for detection and recognition. These files are excluded from Git due to size. Run these commands in the project root: ```Bash chmod +x setup.sh ./setup.sh ``` Or try to do it manually by running the following commands. ```Bash # Face Detection (YuNet) curl -L github.com -o face_detection_yunet_2023mar.onnx # Face Recognition (SFace) curl -L github.com -o face_recognition_sface_2021dec.onnx ``` ## Running the API Start the Server ```Bash cargo run --release ``` The API will be ava …